The impact of climate-change induced temperature increases on the suitability of street tree species in California (USA) cities

Highlights

• Climate change effects on urban tree population assessed with space-for-time substitution.

• Of 16 cities examined, 7 inland ones may suffer substantial species losses.

• Between 55 and 82 of the 140 species found may become locally unsuitable in some cities.

• Climate envelopes model fewer trees as unsuitable than space-for-time substitution.

• Space-for-time substitution is a novel, potentially useful tool in urban forestry.

Abstract

Climate change, resulting in increased temperatures, has the potential to alter the species composition of urban tree populations. We examined the likely changes in composition of common street tree species in California (USA), using a space-for-time substitution approach which paired each of the 16 cities covering the climatic range of the state (“representative city”) with a “warm city” counterpart, where the climate of today approximates the climate of the representative city in 2099. Of the 140 tree species found, as many as 82 species may be unsuitable for the future warmer climate. This change is geographically non-uniform, with greater losses (up to 100% of common species unsuitable for future climate) found in cities away from the ocean coast. In contrast, assessing climate suitability of tree species from published sources (professional judgement, resulting in implied climate envelopes), reduces the number of unsuitable species to as few as 14, although a hybrid approach that accounts for the substantial gaps in the climate envelope information results in an estimated 55 unsuitable tree species. The difference between the observed and estimated results (82 unsuitable species vs. 55) is likely caused by the climate envelope information insufficiently accounting for the irrigation needs of newly-planted street trees in the Mediterranean-type climates like California. Our results demonstrate the viability of the space-for-time substitution approach for evaluating possible climate change effects on urban trees, and suggest both an immediate need to re-evaluate the planting palette of street trees, and a long-term imperative to trial new tree species.

Introduction

Anthropogenic climate change (“global warming”) is contributing to higher temperatures in urban areas, and in temperate climates these increases are especially notable during summers. In 2016, the globally averaged temperature over land and ocean surfaces was the highest since records began in 1880, and the warmest July in 136 years of modern record-keeping was recorded (NOAA, 2017). In California (USA), the average maximum temperature has increased over the past century, with the increase ranging from 0.9 °C to 1.4 °C (1.6–2.5 °F; Lynn, 2015). Projections of climate warming in California indicate further increases in average July average maximum temperatures, projected to range from 2.5 °C (4.6 °F) in Santa Ana, a coastal city in Southern California, to 7.2 °C (13 °F) in the inland city of Barstow (Cal-adapt, 2014). Along with these increasing temperatures, models project a decrease in average annual precipitation and snow accumulation in parts of California, (Diffenbaugh et al., 2015) ranging from -20.3 mm (-0.8 in.) in the northern city of Ukiah, to -162.6 mm (-6.4 in.) in the central-interior city of Stockton (Knowles et al., 2006). These projected reductions are substantial, as they would represent losses ranging from 1% to 42.7% of the current precipitation, respectively (Cal-adapt, 2014).

These changes are potentially impactful on urban environments, as most of California’s population and large cities (e.g., Los Angeles, San Diego, San Jose, Sacramento) are located in Mediterranean-type climates. Characterized by warm-to-hot dry summers, and mild and wet winters (Koppen types Csa, Csb, and Csc; Kottek et al., 2006), the climates of the southern half of California – and the largest cities of Los Angeles, San Diego, and San Jose (combined population of about 20 million) – are also characterized by their potential evapotranspiration exceeding their annual precipitation (e.g., Sanford and Selnick, 2013). Yet, in contrast to the natural vegetation that is xeric or drought-tolerant in the southern and central parts of the state, the current street tree palette across California cities includes many temperate-climate trees (e.g., Ritter, 2011; Thompson, 2006). Some of these urban tree species may be only partially adapted to the current climate, and may require supplemental water, either during their establishment phase, or throughout their life. Thus, a climate with increasing temperatures and decreasing precipitation is likely to have substantial, negative impacts on commonly planted street tree species in California, depending on each species’ adaptation to these new environmental conditions. Physiological changes such as early defoliation and reduced growth can be expected for some species, while others may suffer mortality or fail to break dormancy in spring (Coder, 1999; Tadewaldt, 2013). In turn, these responses by individual trees would be reflected as composition changes in street tree populations. Potential for this effect was observed by the authors during the recent 2013–2016 drought, when below-average precipitation accompanied by reduced availability of water for irrigation of street trees resulted in visible deterioration in street tree condition. This decline was notably varied, both spatially across the state, and across tree species within individual cities, suggesting that the that effects of climate change on street tree populations might be strongly influenced by the interactions of local climate (and its projected change) with the local composition of street tree species, at least in California.

Concerns over such effects of global warming have led to efforts to estimate the suitability of ornamental landscape species, including trees, to future climates and to provide planting guidance to practitioners. For example, a plant heat zone map for the United States was created by the American Horticultural Society (American Horticultural Society (AHS, 1997; LaLiberte, 2016), with heat zones defined by the yearly number of “heat days” (daily max. temp >30 °C or 86 °F). However, the listed examples of plants that perform well in each heat zone do not include commonly used street trees. Conversely, Buley and Cregg (2016) stressed the need for urban tree selection with a view toward the changing climate, and suggested – based on the general understanding of species’ physiology and phenology – increased use of species from more phenotypically plastic genera such as Acer, Gymnocladus, Nyssa, Plantanus, Taxodium, and Ulmus. Similarly, in California, a limited number of publications include professional opinions on the adaptation of trees to California landscape regions (Brenzel, 2007; Perry, 2010; Selectree, 2016), and professional consensus on the relative water requirements of common landscape plants (Costello and Jones, 2014). While these publications represent valuable sources of information for judging the adaptation of tree species to a warming and drying climate, they do not substitute for systematic evaluation of tree performance. At least one such study that includes growing novel species and cultivars and evaluating their performance in several cities across California is currently in progress (McPherson et al., 2018), but the results will not be known for many years.

Alternative approaches have been proposed for evaluating plant performance in changing climate and predicting effects of climate change on natural ecosystems (e.g., Anacker et al., 2013). Broadmeadow and colleagues proposed in 2005 that to maintain timber production in a changing climate, “provenances from regions with a current climate similar to that predicted for the future for a given site should be selected” (p. 157), and in 2013 Blois et al. suggested that “space-for-time substitution” method, commonly used in studies of plant succession (Pickett, 1989), may be useful in predicting climate-change effects on biodiversity. In urban forests, a range of approaches to evaluating the effect of climate change on tree populations has been used. These include relatively simple approaches based on the trees’ “climate envelope” (i.e., the set of climatic constraints on a species’ distribution) focusing on climate variables and projected pest problems in a changing climate (e.g., Yang, 2009). But the approaches also include increasingly more complex analyses, such as when various management considerations are explicitly included (e.g., Ordóñez and Duinker, 2014, 2015), and becoming even more involved when social considerations and larger-scale constraints are introduced (e.g., Brandt et al., 2016). Brandt et al. (2017) applied the framework they developed (2016) to the urban forests in the Chicago region. Their analysis combined a modeled climate envelope for native trees, together with a tolerance of heat and cold for non-native trees (as indicated by AHS Heat Zones and USDA Hardiness Zones) with assigned scores for adaptability to pests and diseases. The results of their analysis produced two vulnerability categories (low; high) of tree species in the Chicago area to climate change. Yet, none of the studies have been carried out in Mediterranean-type climates, where decreased water and increased summer heat are likely to be significant constraints in the changing climate.

This work presents the first use of the space-for-time substitution method to evaluate the potential impact of climate change on street tree populations. We assess the composition of common street tree species in California cities, to identify the common species that would be expected to perform poorly or would fail to survive in the warmer climate. We also briefly compare the output of the space-for-time substitution to the results of professional opinion in predicting the future suitability of street tree species.